Connectedness & Cohesion¶

Definition & Conceptual Goal¶

The Connectedness & Cohesion metric evaluates whether a theory-net maintains a unified, non-anarchical logical lineage and measures its resilience against fragmentation. In structuralist philosophy of science (Balzer et al., 1987, p. 173), a scientific theory is not a random collection of disjoint laws; all specialized laws must logically connect back to common theoretical roots.

Beyond basic topological reachability, cohesion measures evaluate the Quinean Web of Belief (Quine, 1951)—how well the theory absorbs empirical shocks and whether it suffers from severe bottlenecks or weak links that could shatter the framework.

Theoretical Grounding & Model Formulation¶

Let a theory-net be defined as: $$ N = \langle T, \alpha \rangle $$

where \(T = \{T_1, T_2, \dots, T_n\}\) is a finite set of theory-elements \(T_i = \langle K_i, I_i \rangle\), and \(\alpha\) represents specialization relations.

While a theory may be technically connected (a single component), it might be highly fragile. We define three advanced levels of cohesion:

1. \(k\)-connectivity (Vertex Robustness): The absolute weakest link. How many theoretical nodes must be falsified to split the theory into isolated parts?
2. Max-Flow Min-Cut (Edge Robustness): How many relational claims (edges) must be severed to disconnect the theory?
3. Algebraic Connectivity (\(\lambda_2\)): The continuous measure of global diffusion. It dictates how quickly a random walk mixes across the network, modeling how efficiently the Quinean web can share the "burden of adjustment" across distant propositions when anomalies arise.

Mathematical Specification & Graph Formulation¶

Let \(G = (V, E)\) be the undirected version of the theory graph.

Basic Connectedness (Reachability)¶

$$ \text{Connectedness} (G) = \begin{cases} 1 & \text{if } |\pi_0 (G)| = 1 \ 0 & \text{otherwise} \end{cases} $$ where \(|\pi_0 (G)|\) is the number of connected components.

Vertex & Edge Connectivity (\(k\)-connectivity & Min-Cut)¶

• Vertex Connectivity \(\kappa (G)\): The minimum number of nodes whose removal disconnects \(G\). Represents "Hard Falsifiability."
• Edge Connectivity \(\lambda (G)\): The minimum number of edges whose removal disconnects \(G\). By the Max-Flow Min-Cut theorem, this is equivalent to the maximum flow between the most weakly connected domains. Always: \(\kappa (G) \le \lambda (G) \le \delta (G)\) (where \(\delta\) is the minimum degree).

Algebraic Connectivity (\(\lambda_2\))¶

Let \(L = D - A\) be the graph Laplacian, where \(D\) is the degree matrix and \(A\) is the adjacency matrix. The eigenvalues of \(L\) are \(0 = \lambda_1 \le \lambda_2 \le \dots \le \lambda_n\). The second smallest eigenvalue, \(\lambda_2\) (Fiedler value), is the Algebraic Connectivity.

• \(\lambda_2 > 0\) if and only if the graph is connected.
• Higher \(\lambda_2\) indicates a robust, highly integrated graph with no severe bottlenecks.

Measurement & Graph Implementation¶

1. WCC: Run weakly connected components (WCC) to test basic reachability.
2. Min-Cut / Max-Flow: Compute the minimum edge cut between major theory modules to find relational weak points.
3. Laplacian Spectrum: Calculate \(\lambda_2\) to determine the global integration score and identify the Fiedler vector (which partitions the graph along its worst bottleneck).

Diagnostic & Metascientific Value¶

Comparing continuous Algebraic Connectivity (\(\lambda_2\)) against discrete \(k\)-connectivity provides a precise epistemic profile of the theory:

Grounding References¶

• [Balzer et al., 1987] Balzer, W., Moulines, C. U., & Sneed, J. D. (1987). An Architectonic for Science. Reidel Publishing Company.
• [Quine, 1951] Quine, W. V. O. (1951). Two Dogmas of Empiricism. The Philosophical Review, 60 (1), 20–43.